Cancer consists of a group of diseases which involve unregulated cell growth and death, genome instability and mutations, tumor-promoting inflammation, induction of angiogenesis, immune system evasion, deregulation of metabolic pathways, immortal cell replication, and metastatic tissue invasion [1]. Cancer is the second leading cause of death in the United States after heart disease [2]. More than 1.6 million new cases of cancer are projected to be diagnosed each year, with more than 580,000 Americans expected to die (about 1600 cancer deaths per day), accounting for nearly 1 in 4 of all American deaths [2, 3].
The immune system plays an important role in the development and progression of cancer. Immune cell infiltration to the tumor site can adversely affect malignancy progression and metastasis [4, 5]. Infiltration of macrophages into the tumor site has been shown to account for more than 50% of the tumor mass in certain breast cancer cases suggesting macrophages have a significant role in tumor progression [6-8].
Macrophages are cells derived from the myeloid lineage and belong to the innate immune system. They are derived from blood monocytes that migrate into tissue. One of their main functions is to phagocytose microbes and clear cellular debris. They also play an important role in both the initiation and resolution of inflammation [9, 10]. Moreover, macrophages can display different responses, ranging from pro-inflammatory to anti-inflammatory, depending on the type of stimuli they receive from the surrounding microenvironment [11]. Two major macrophage phenotypes have been proposed which correlate with extreme macrophage responses: M1 and M2.
M1 pro-inflammatory macrophages are activated upon contact with certain molecules such as lipopolysaccharide (LPS), IFN-γ, IL-1β, TNF-α, and Toll-like receptor engagement. M1 macrophages constitute a potent arm of the immune system deployed to fight infections. They are capable of either direct (pathogen pattern recognition receptors) or indirect (Fc receptors, complement receptors) recognition of the pathogen. They are also armed in their ability to produce reactive oxygen species (ROS) as means to help killing pathogens. In addition, M1 macrophages secrete pro-inflammatory cytokines and chemokines attracting other types of immune cells and integrating/orchestrating the immune response. M1 activation is induced by IFN-g, TNFa, GM-CSF, LPS and other toll-like receptors (TLR) ligands.
In contrast, M2 anti-inflammatory macrophages, also known as alternatively activated macrophages, are activated by anti-inflammatory molecules such as IL-4, IL-13, and IL-10 [12, 13]. M2 macrophages exhibit immunomodulatory, tissue repair, and angiogenesis properties which allow them to recruit regulatory T cells to sites of inflammation. M2 macrophages do not constitute a uniform population and often are further subdivided into M2a, M2b and M2c categories. The common denominator of all three subpopulations is high IL-10 production accompanied by low production of IL-12. One of their signatures is production of enzyme Arginase-1 that depletes L-arginine thereby suppressing T cell responses and depriving iNOS of its substrate.
The in vivo molecular mechanisms of macrophage polarization are poorly characterized because of the variety of signals macrophages experience in the cellular microenvironment [10, 14]. In recent years, progress has been made in identifying in vivo macrophage polarization under physiological conditions such as ontogenesis, pregnancy, and pathological conditions such as allergies, chronic inflammation, and cancer. We do know, however, that in vitro macrophage polarization is plastic and macrophages, with the help of cytokines, can be polarized back and forth to either phenotype [15, 16]. Interferon gamma (IFN-γ) and IL-4 are two cytokines that can polarize macrophages to M1 and M2 phenotypes, respectively [15].
The presence of macrophages is crucial for tumor progression and growth, and has implications in determining prognosis [17, 18]. Because macrophages can exhibit both pro-inflammatory and anti-inflammatory properties, it is important to understand their polarization and function in tumor progression and metastasis.
Macrophage Polarization
The tumor microenvironment can affect macrophage polarization. The process of polarization can be diverse and complex because of the hostile environment of IL-10, glucocorticoid hormones, apoptotic cells, and immune complexes that can interfere with innate immune cells function [11, 19]. The mechanisms of polarization are still unclear but we know they involve transcriptional regulation. For example, macrophages exposed to LPS or IFN-γ will polarize towards an M1 phenotype, whereas macrophages exposed to IL-4 or IL-13 will polarize towards an M2 phenotype. LPS or IFN-γ can interact with Toll-like receptor 4 (TLR4) on the surface of macrophages inducing the Trif and MyD88 pathways, inducing the activation of transcription factors IRF3, AP-1, and NFκB and thus activating TNFs genes, interferon genes, CXCL10, NOS2, IL-12, etc., which are necessary in a pro-inflammatory M1 macrophage response [20]. Similarly, IL-4 and IL-13 bind to IL-4R, activation the Jak/Stat6 pathway, which regulates the expression of CCL17, ARG1, IRF4, IL-10, SOCS3, etc., which are genes associated with an anti-inflammatory response (M2 response).
Additional mechanisms of macrophage polarization include microRNA (miRNA) micromanagement. miRNAs are small non-coding RNA of 22 nucleotides in length that regulate gene expression post-transcriptionally, as they affect the rate of mRNA degradation. Several miRNAs have been shown to be highly expressed in polarized macrophages, especially miRNA-155, miRNA-125, miRNA-378 (M1 polarization), and miRNA let-7c, miRNA-9, miRNA-21, miRNA-146, miRNA147, miRNA-187 (M2 polarization) [21].
Macrophage polarization is a complex process, were macrophages behave and elicit different responses depending on microenvironment stimuli. Therefore, macrophage polarization is better represented by a continuum of activation states where M1 and M2 phenotypes are the extremes of the spectrum. In recent years, there has been much controversy on the definition/description of macrophage activation and macrophage polarization. A recent paper published by Murray et al., in which they describe a set of standards to be considered for the consensus definition/description of macrophage activation, polarization, activators, and markers. This publication was much needed for the definition and characterization of activated/polarized macrophages [22].
M1 Phenotype
M1 pro-inflammatory macrophages or classically activated macrophages are aggressive, highly phagocytic, and produce large amounts of reactive oxygen and nitrogen species, thereby promoting a Th1 response [11]. M1 macrophages secrete high levels of two important inflammatory cytokines, IL-12 and IL-23. IL-12 induces the activation and clonal expansion of Th17 cells, which secrete high amounts of IL-17, which contributes to inflammation [23]. These characteristics allow M1 macrophages to control metastasis, suppress tumor growth, and control microbial infections [24]. Moreover, the infiltration and recruitment of M1 macrophages to tumor sites correlates with a better prognosis and higher overall survival rates in patients with solid tumors [17, 18, 25-28].
Polarization of macrophages to the M1 phenotype is regulated in vitro by inflammatory signals such as IFN-γ, TNF-α, IL-1β and LPS as well as transcription factors and miRNAs [29, 30]. Classically activated macrophages initiate the induction of the STAT1 transcription factor which targets CXCL9, CXCL10 (also known as IP-10), IFN regulatory factor-1, and suppressor of cytokine signaling-1 [31]. Cytokine signaling-1 protein functions downstream of cytokine receptors, and takes part in a negative feedback loop to attenuate cytokine signaling. In the tumor microenvironment, Notch signaling plays an important role in the polarization of M1 macrophages, as it allows transcription factor RBP-J to regulate classical activation. Macrophages that are deficient in Notch signaling express an M2 phenotype regardless of other extrinsic inducers [32]. One crucial miRNA, miRNA-155, is upregulated when macrophages are transitioning from M2 to M1; M1 macrophages overexpressing miRNA-155 are generally more aggressive and are associated with tumor reduction [33]. Moreover, miRNA-342-5p has been found to foster a greater inflammatory response in macrophages by targeting Akt1 in mice. This miRNA also promotes the upregulation of Nos2 and IL-6, both of which act as inflammatory signals for macrophages [34]. Other miRNAs such as miRNA-125 and miRNA-378 have also been shown to be involved in the classical activation pathway of macrophages (M1) [35].
Classically activated macrophages are thought to play an important role in the recognition and destruction of cancer cells as their presence usually indicates good prognosis. After recognition, malignant cells can be destroyed by M1 macrophages through several mechanisms, which include contact-dependent phagocytosis and cytotoxicity (i.e., cytokine release such as TNF-α) [24]. Environmental signals such as the tumor microenvironment or tissue-resident cells, however, can polarize M1 macrophages to M2 macrophages. In vivo studies of murine macrophages have shown that macrophages are plastic in their cytokine and surface marker expression and that re-polarizing macrophages to an M1 phenotype in the presence of cancer can help the immune system reject tumors [19].
M2 Phenotype
M2 macrophages are anti-inflammatory and aid in the process of angiogenesis and tissue repair. They express scavenger receptors and produce large quantities of IL-10 and other anti-inflammatory cytokines [33, 36]. Expression of IL-10 by M2 macrophages promotes a Th2 response. Th2 cells consequently upregulate the production of IL-3 and IL-4. IL-3 stimulates proliferation of all cells in the myeloid lineage (granulocytes, monocytes, and dendritic cells), in conjunction with other cytokines, e.g., Erythropoietin (EPO), Granulocyte macrophage colony-stimulating factor (GM-CSF), and IL-6. IL-4 is an important cytokine in the healing process because it contributes to the production of the extracellular matrix [23]. M2 macrophages exhibit functions that may help tumor progression by allowing blood vessels to feed the malignant cells and thus promoting their growth. The presence of macrophages (thought to be M2) in the majority of solid tumors negatively correlates with treatment success and longer survival rates [37]. Additionally, the presence of M2 macrophages has been linked to the metastatic potential in breast cancer. Lin and colleagues found that early recruitment of macrophages to the breast tumor sites in mice increase angiogenesis and incidence of malignancy [38]. It is thought that the tumor microenvironment helps macrophages maintain an M2 phenotype [23, 39]. Anti-inflammatory signals present in the tumor microenvironment such as adiponectin and IL-10 can enhance an M2 response [41].
Tumor-Associated Macrophages (TAMs)
Cells exposed to a tumor microenvironment behave differently. For example, tumor-associated macrophages found in the periphery of solid tumors are thought to help promote tumor growth and metastasis, and have an M2-like phenotype [42]. Tumor-associated macrophages can be either tissue resident macrophages or recruited macrophages derived from the bone marrow (macrophages that differentiate from monocytes to macrophages and migrate into tissue). A study by Cortez-Retamozo found that high numbers of TAM precursors in the spleen migrate to the tumor stroma, suggesting this organ as a TAM reservoir also [43]. TAM precursors found in the spleen were found to initiate migration through their CCR2 chemokine receptor [43]. Recent studies have found CSF-1 as the primary factor that attracts macrophages to the tumor periphery, and that CSF-1 production by cancer cells predicts lower survival rates and it indicates an overall poor prognosis [44-46]. Other cytokines such as TNF-α and IL-6 have been also linked to the accumulation/recruitment of macrophages to the tumor periphery [45].
It is thought that macrophages that are recruited around the tumor borders are regulated by an “angiogenic switch” that is activated in the tumor. The angiogenic switch is defined as the process by which the tumor develops a high density network of blood vessels that potentially allows the tumor to become metastatic, and is necessary for malignant transition. In a breast cancer mouse model, it was observed that the presence of macrophages was required for a full angiogenic switch. When macrophage maturation, migration, and accumulation around the tumor was delayed, the angiogenic switch was also delayed suggesting that the angiogenic switch does not occur in the absence of macrophages and that macrophage presence is necessary for malignancy progression [47]. Moreover, the tumor stromal cells produce chemokines such as CSF1, CCL2, CCL3, CCL5, and placental growth factor that will recruit macrophages to the tumor surroundings. These chemokines provide an environment for macrophages to activate the angiogenic switch, in which macrophages will produce high levels of IL-10, TGF-β, ARG-1 and low levels of IL-12, TNF-α, and IL-6. The level of expression of these cytokines suggests macrophages modulate immune evasion. It is important to note that macrophages are attracted to hypoxic tumor environments and will respond by producing hypoxia-inducible factor-1α (HIF-1α) and HIF-2α, which regulate the transcription of genes associated with angiogenesis. During the angiogenic switch, macrophages can also secrete VEGF (stimulated by the NF-κB pathway), which will promote blood vessel maturation and vascular permeability [48].
Tumor-associated macrophages are thought to be able to maintain their M2-like phenotype by receiving polarization signals from malignant cells such as IL-1R and MyD88, which are mediated through IkB kinase β and NF-kB signaling cascade. Inhibition of NF-kB in TAMs promotes classical activation [40]. Moreover, another study suggested that p50 NF-kB subunit was involved in suppression of M1 macrophages, and reduction of inflammation promoted tumor growth. A p50 NF-κB knock-out mouse generated by Saccani et al. suggested that M1 aggressiveness was restored upon p50 NF-kB knockout, reducing tumor survival [49].
Because the tumor mass contains a great number of M2-like macrophages, TAMs can be used as a target for cancer treatment. Reducing the number of TAMs or polarizing them towards an M1 phenotype can help destroy cancer cells and impair tumor growth [50-52]. Luo and colleagues used a vaccine against legumain, a cysteine protease and stress protein upregulated in TAMs thought to be a potential tumor target [52]. When the vaccine against legumain was administered to mice, genes controlling angiogenesis were downregulated and tumor growth was halted [52].
Metabolism and Activation Pathways
Metabolic alterations present in tumor cells are controlled by the same genetic mutations that produce cancer [53]. As a result of these metabolic alterations, cancer cells are able to produce signals that can modify the polarization of macrophages and promote tumor growth [54, 55].
M1 and M2 macrophages demonstrate distinct metabolic patterns that reflect their dissimilar behaviors [56]. The M1 phenotype increases glycolysis and skews glucose metabolism towards the oxidative pentose phosphate pathway, thereby decreasing oxygen consumption and consequently producing large amounts of radical oxygen and nitrogen species as well as inflammatory cytokines such as TNF-α, IL-12, and IL-6 [56, 57]. The M2 phenotype increases fatty acid intake and oxidation, which decreases flux towards the pentose phosphate pathway while increasing the overall cell redox potential, consequently upregulating scavenger receptors and immunomodulatory cytokines such as IL-10 and TGF-β [56].
Multiple metabolic pathways play important roles in macrophage polarization. Protein kinases, such as Akt1 and Akt2, alter macrophage polarization by allowing cancer cells to survive, proliferate, and use an intermediary metabolism [58]. Other protein kinases can direct macrophage polarization through glucose metabolism by increasing glycolysis and decreasing oxygen consumption [57, 59]. Shu and colleagues were the first to visualize macrophage metabolism and immune response in vivo using a PET scan and a glucose analog [60].
L-arginine metabolism also exhibits discrete shifts important to cytokine expression in macrophages and exemplifies distinct metabolic pathways which alter TAM-tumor cell interactions [61]. Classically activated (M1) macrophages favor inducible nitric oxide synthase (iNOS). The iNOS pathway produces cytotoxic nitric oxide (NO), and consequently exhibits anti-tumor behavior. Alternatively activated (M2) macrophages have been shown to favor the arginase pathway and produce ureum and 1-ornithine, which contribute to progressive tumor cell growth [61, 62].
Direct manipulation of metabolic pathways can alter macrophage polarization. The carbohydrate kinase-like protein (CARKL) protein, which plays a role in glucose metabolism, has been used to alter macrophage cytokine signatures [56, 57]. When CARKL is knocked down by RNAi, macrophages tend to adopt an M1-like metabolic pathway (metabolism skewed towards glycolysis and decreased oxygen consumption). When CARKL is overexpressed, macrophages adopt an M2-like metabolism (decreased glycolytic flux and more oxygen consumption) [56]. When macrophages adopt an M1-like metabolic state through LPS/TLR4 engagement, CARKL levels decrease, genes controlled by the NFκB pathway are activated (TNF-α, IL-12, and IL-6), and cell redox potential increases due to growing concentrations of NADH:NAD+ and GSH:GSSSG complexes. During an M2-like metabolic state, macrophages upregulate CARKL and genes regulated by STAT6/IL-4 (IL-10 and TGF-β).
Obesity can also affect macrophage polarization. Obesity is associated with a state of chronic inflammation, an environment that drives the IL4/STAT6 pathway to activate NKT cells, which drive macrophages towards an M2 response. During late-stage diet-induced obesity, macrophages migrate to adipose tissue, where immune cells alter levels of TH1 or TH2 cytokine expression in the adipose tissue, causing an M2 phenotype bias and possibly increased insulin sensitivity [63].
M1 phenotype bias by targeting metabolic pathways in TAMS may offer an alternative means of reducing tumor growth and metastasis.
Macrophage Immunotherapy Approaches Against Cancer
The role of cancer immunotherapy is to stimulate the immune system to recognize, reject, and destroy cancer cells. Cancer immunotherapy with monocytes/macrophages has the goal to polarize macrophages towards a pro-inflammatory response (M1), thus allowing the macrophages and other immune cells to destroy the tumor. Many cytokines and bacterial compounds can achieve this in vitro, although the side effects are typically too severe in vivo. The key is to find a compound with minimal or easily managed patient side effects. Immunotherapy using monocytes/macrophages has been used in past decades and new approaches are being developed every year [64, 65]. Early immunotherapy has established a good foundation for better cancer therapies and increased survival rate in patients treated with immunotherapies [66].
Some approaches to cancer immunotherapy include the use of cytokines or chemokines to recruit activated macrophages and other immune cells to the tumor site which allow for recognition and targeted destruction of the tumor site [67, 68]. IFN-α and IFN-β have been shown to inhibit tumor progression by inducing cell differentiation and apoptosis [69]. Also, IFN treatments are anti-proliferative and can increase S phase time in the cell cycle [70, 71]. Zhang and colleagues performed a study in nude mice using IFN-β gene therapy to target human prostate cancer cells. Their results indicate that adenoviral-delivered IFN-β gene therapy involves macrophages and helps suppress growth and metastasis [72].
The macrophage inhibitory factor (MIF) is another cytokine that can be used in cancer immunotherapy. MIF is usually found in solid tumors and indicates poor prognosis. MIF inhibits aggressive macrophage function and drives macrophages toward an M2 phenotype, which can aid tumor growth and progression. Simpson, Templeton & Cross (2012) found that MIF induces differentiation of myeloid cells, macrophage precursors, into a suppressive population of myeloid cells that express an M2 phenotype [73]. By targeting MIF, they were able to deplete this suppressive population of macrophages, inhibiting their growth and thus control tumor growth and metastasis [73].
The chemokine receptor type 2, CCR2, is crucial to the recruitment of monocytes to inflammatory sites and it has been shown as a target to prevent the recruitment of macrophages to the tumor site, angiogenesis, and metastasis. Sanford and colleagues (2013) studied a novel CCR2 inhibitor (PF-04136309) in a pancreatic mouse model, demonstrating that the CCR2 inhibitor depleted monocyte/macrophage recruitment to the tumor site, decreased tumor growth and metastasis, and increased antitumor immunity [74]. Another recent study by Schmall et al. showed that macrophages co-cultured with 10 different human lung cancers upregulated CCR2 expression. Moreover, they showed that tumor growth and metastasis were reduced in a lung mouse model treated with a CCR2 antagonist [75].
Other studies have used liposomes to deliver drugs to deplete M2 macrophages from tumors and to stop angiogenesis. Cancer cells that express high levels of IL-1β grow faster and induce more angiogenesis in vivo. Kimura and colleagues found that macrophages exposed to tumor cells expressing IL-1β produced higher levels of angiogenic factors and chemokines such as vascular endothelial growth factor A (VEG-A), IL-8, monocyte chemoattractant protein 1, etc., facilitating tumor growth and angiogenesis [76]. When they used clodronate liposomes to deplete macrophages, they found fewer IL-10-producing tumor cells. They also found that by inhibiting NF-κB and AP-1 transcription factors in the cancer cells, tumor growth and angiogenesis were reduced. These findings may suggest that macrophages that surround the tumor site may be involved in promoting tumor growth and angiogenesis [76].
Compounds such as methionine enkephalin (MENK) have anti-tumor properties in vivo and in vitro. MENK has the ability to polarize M2 macrophages to M1 macrophages by downregulating CD206 and arginase-1 (M2 markers) while upregulating CD64, MHC-II, and the production of nitric oxide (M1 markers). MENK can also upregulate TNF-α and downregulate IL-10 [77].
Recent studies have focused on bisphosphonates as a potential inhibitor of M2 macrophages. Bisphosphonates are commonly used to treat metastatic breast cancer patients to prevent skeletal complications such as bone resorption [78]. While bisphosphonates stay in the body for short periods of time, bisphosphonates can target osteoclasts, cells in the same family as macrophages, due to their high affinity for hydroxyapatite. Once bisphosphonates bind to the bones, the bone matrix internalizes the bisphosphonates by endocytosis. Once in the cytoplasm, bisphosphonates can inhibit protein prenylation, an event that prevents integrin signaling and endosomal trafficking, thereby forcing the cell to go apoptotic [69]. Until recently, it was unknown whether bisphosphonates could target tumor-associated macrophages but a recent study by Junankar et al. has shown that macrophages uptake nitrogen-containing bisphosphonate compounds by pinocytosis and phagocytosis, an event that does not occur in epithelial cells surrounding the tumor [79]. Forcing TAMs to go apoptotic using bisphosphonates could reduce angiogenesis and metastasis.
Additional approaches to cancer immunotherapy include the use of biomaterials that may elicit an immune response. Cationic polymers are used in immunotherapy because of their reactivity once dissolved in water. Chen et al. used cationic polymers including PEI, polylysine, cationic dextran and cationic gelatin to produce a strong Th1 immune response [77]. They were also able to induce proliferation of CD4+ cells and secretion of IL-12 typical of M1 macrophages [77]. Huang and colleagues also used biomaterials to trigger TAMs to produce an anti-tumor response by targeting TLR4 [80]. This study found that TAMs were able to polarize to an M1 phenotype and express IL-12. They found that these cationic molecules have direct tumoricidal activity and demonstrate tumor reduction in mice [80].
TLR4
Toll-like receptor 4 is a protein in humans that is encoded by the TLR4 gene. TLR 4 detects lipopolysaccharide (LPS) on gram negative bacteria and thus plays a fundamental role in the recognition of danger and the activation of the innate immune system (FIG. 7). It cooperates with LY96 (MD-2) and CD14 to mediate signal transduction when macrophages are induced by LPS. The cytoplasmic domain of TLR4 is responsible for the activation of M1 macrophages when they detect the presence of LPS. This is the functional portion of the receptor that would be coupled to the MOTO-CAR (i.e., chimeric receptor) to induce activation of the monocyte/macrophage when the CAR binds its target protein.
The adaptor proteins MyD88 and TIRAP contribute to the activation of several and possibly all pathways via direct interactions with TLR4's Toll/interleukin-1 receptor (IL-1R) (TIR) domain. However, additional adaptors that are required for the activation of specific subsets of pathways may exist, which could contribute to the differential regulation of target genes.
Thymidine Kinase
Human Thymidine Kinase 1 (TK1) is a well-known nucleotide salvage pathway enzyme that has largely been studied in the context of its overexpression in tumors. Since TK1 was initially popularized by its expression in the serum of cancer patients (sTK), its diagnostic and prognostic potential has been studied extensively. For example, several studies have demonstrated that sTK1 in many different cancer patients is elevated in a stage-like manner with a higher level of TK1 indicating a more advanced tumor [81].
Other studies have investigated the prognostic potential of TK1. One such study demonstrates that the TK1 levels in primary breast tumors can be used to predict recurrence. Other exciting TK1 prognostic studies show significant reductions in sTK1 levels when patients respond to treatment while sTK1 levels continue to rise in patients who do not appear to respond to their treatment. It is also known that sTK1 levels begin to rise prior to recurrence and noted in some cases sTK1 levels could predict recurrence “1-6 months before the onset of clinical symptoms.” Several other studies confirm the rich potential of TK1 as a diagnostic and prognostic indicator of cancer [82].
Although the diagnostic and prognostic potential of TK1 has been well established, the therapeutic potential of TK1 remains veiled in comparison. While it is true that HSV-TK has been used in gene therapy and PET imaging utilizes TK1 to identify proliferating cancer cells, few, if any studies address the possibility of a TK1 immunotherapy. Perhaps this is primarily because TK1 is a known cytosolic protein. It has been recently discovered that TK1 is expressed not only in cancer cells but also on the surface membrane of multiple tumor types and is therefore a very viable target for tumor immunotherapy.